A method for quantization scale factor selection in MPEG2 videosequence encoding using a bilayer fuzzy expert system

摘要

This paper presents a method for intelligent selection of thequantization scale factor used for varying bit rate and quality in theMPEG2 video encoding standard. The goal is to select the quantizationscale factor, above which, noticeable quality degradation occurs. Thismethod uses a bilayer fuzzy logic system to aggregate measurements takenon the block level. The rulebases are designed around thecharacteristics of the human visual system (HVS). By taking advantage ofthe HVS characteristics, an informed selection of the quantization scalefactor with respect to the resulting quality can be made